[gentoo-user] Hard drive RPMs and data speed.

[gentoo-user] Hard drive RPMs and data speed.

[Dale]

Howdy,

I'm wanting to get a hard drive that is pretty good size.  I'm looking 
for about 1 to 2TBs or so.  Thing is, a lot of them seem to be 5900 or 
even 5400 rpm drives.  I realize that the data on there is packed pretty 
tight so I want to ask a few people that may have one or more of these 
things a few questions.  Are they as fast as a slower RPM drive?  Would 
they be fast enough to play HD videos and such?  I have quite a few 1080 
HD videos.  I don't want the drive to cause issues.

For additional info, I put my videos on their own drive.  I have a 750Gb 
drive that is mounted on /data and videos are placed in there.  When I 
am playing a video, the only thing being accessed on that drive is the 
video.  My OS, /home partition and such is on a separate drive.  The 
drives are also SATA 3Gbs/sec drives.  My mobo doesn't have SATA 6Gbs/sec.

Can someone that has one or more of these post their hdparm -Tt 
results?  Different speeds would be great too. I'd like to compare what 
a 5400rpm drive would do compared to a 7200rpm drive.

Thoughts and info please.

Dale

:-)  :-)

[gentoo-user] Re: Hard drive RPMs and data speed.

[Nikos Chantziaras]

On 10/27/2011 11:15 AM, Dale wrote:
> Howdy,
>
> I'm wanting to get a hard drive that is pretty good size. I'm looking
> for about 1 to 2TBs or so. Thing is, a lot of them seem to be 5900 or
> even 5400 rpm drives. I realize that the data on there is packed pretty
> tight so I want to ask a few people that may have one or more of these
> things a few questions. Are they as fast as a slower RPM drive?

I assume you meant to say "as fast as a faster RPM drive".  No, of 
course not.  If we're speaking about the same capacity and amount of 
platters, of course.  If we're not, then yes, they can be as fast 
because of the higher data density.


> Would
> they be fast enough to play HD videos and such? I have quite a few 1080
> HD videos. I don't want the drive to cause issues.

The transfer speed required for playing HD videos is virtually zero. 
1080p video compressed using an 8mbps rate require 2MB/s.  This can be 
done even with the slowest drive from 10 years ago.  Today's slowest 
drive are able to play about 40 or 50 of those HD video simultaneously. 
  So the answer is yes.  They can play HD video :-)

Most of those 5900/5400 disks are meant for pure data storage.  The 
lower RPM is used to market them as "green and silent", meaning they 
don't consume much power and aren't noisy.  Installing your OS on them 
though isn't going to give you good speed.  They have good transfer 
rates, but their access times usually suck.


> Can someone that has one or more of these post their hdparm -Tt results?
> Different speeds would be great too. I'd like to compare what a 5400rpm
> drive would do compared to a 7200rpm drive.

Simply Google around for benchmarks of the drivers you're interested in. 
  Note that is in area where it doesn't make any real difference that 
the benches or reviews you find are performed under MS Windows.  The 
results are applicable to every OS.

As a rule of thumb when buying drives: if you want to install software 
on it, buy an 7200RPM drive with good access times.  Of course they're 
more expensive  If you just want to store all your downloaded HD porn 
and music collection on it, a silent 5400RPM drive is a good choice.

Oh, and one other thing; hdparm is only meant to get you the continuous 
I/O transfer rate.  It's an awful benchmark for anything else, like what 
happens if a file is fragmented or how fast it can copy/write data 
spread around the disk, how good it is at combined random I/O operation, 
etc.

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Thu, Oct 27, 2011 at 6:09 AM, Nikos Chantziaras <realnc@arcor.de> wrote:
> Oh, and one other thing; hdparm is only meant to get you the continuous I/O
> transfer rate.  It's an awful benchmark for anything else, like what happens
> if a file is fragmented or how fast it can copy/write data spread around the
> disk, how good it is at combined random I/O operation, etc.

For that kind of information, go with bonnie++

I've little else to add to the thread, except that I ran three Seagate
1.5TB 'green' drives in RAID5 for quite a while with very nice
perforance results. Access times were comfy, and I tended to get about
60MB/s continuous read and write speed. I hadn't learned about
bonnie++ yet, so I don't have any good benchmarks to show on that
front.

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Donnerstag 27 Oktober 2011, 13:09:17 schrieb Nikos Chantziaras:
> On 10/27/2011 11:15 AM, Dale wrote:
> > Howdy,
> > 
> > I'm wanting to get a hard drive that is pretty good size. I'm looking
> > for about 1 to 2TBs or so. Thing is, a lot of them seem to be 5900 or
> > even 5400 rpm drives. I realize that the data on there is packed pretty
> > tight so I want to ask a few people that may have one or more of these
> > things a few questions. Are they as fast as a slower RPM drive?
> 
> I assume you meant to say "as fast as a faster RPM drive".  No, of
> course not.  If we're speaking about the same capacity and amount of
> platters, of course.  If we're not, then yes, they can be as fast
> because of the higher data density.
> 
> > Would
> > they be fast enough to play HD videos and such? I have quite a few 1080
> > HD videos. I don't want the drive to cause issues.
> 
> The transfer speed required for playing HD videos is virtually zero.
> 1080p video compressed using an 8mbps rate require 2MB/s.  This can be
> done even with the slowest drive from 10 years ago.  Today's slowest
> drive are able to play about 40 or 50 of those HD video simultaneously.
>   So the answer is yes.  They can play HD video :-)
> 
> Most of those 5900/5400 disks are meant for pure data storage.  The
> lower RPM is used to market them as "green and silent", meaning they
> don't consume much power and aren't noisy.  Installing your OS on them
> though isn't going to give you good speed.  They have good transfer
> rates, but their access times usually suck.
> 
> > Can someone that has one or more of these post their hdparm -Tt results?
> > Different speeds would be great too. I'd like to compare what a 5400rpm
> > drive would do compared to a 7200rpm drive.
> 
> Simply Google around for benchmarks of the drivers you're interested in.
>   Note that is in area where it doesn't make any real difference that
> the benches or reviews you find are performed under MS Windows.  The
> results are applicable to every OS.
> 
> As a rule of thumb when buying drives: if you want to install software
> on it, buy an 7200RPM drive with good access times.  Of course they're
> more expensive  If you just want to store all your downloaded HD porn
> and music collection on it, a silent 5400RPM drive is a good choice.
> 

indeed. Additionally they don't get really warm. Which reduces the overall 
thermal load in the case.

One important thing:

most if not all 2TB drives have 4K sectors, which means you have to be 
carefull while partitioning those beasts.
-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Dale]

Volker Armin Hemmann wrote:
> Am Donnerstag 27 Oktober 2011, 13:09:17 schrieb Nikos Chantziaras:
>> On 10/27/2011 11:15 AM, Dale wrote:
>>> Howdy,
>>>
>>> I'm wanting to get a hard drive that is pretty good size. I'm looking
>>> for about 1 to 2TBs or so. Thing is, a lot of them seem to be 5900 or
>>> even 5400 rpm drives. I realize that the data on there is packed pretty
>>> tight so I want to ask a few people that may have one or more of these
>>> things a few questions. Are they as fast as a slower RPM drive?
>> I assume you meant to say "as fast as a faster RPM drive".  No, of
>> course not.  If we're speaking about the same capacity and amount of
>> platters, of course.  If we're not, then yes, they can be as fast
>> because of the higher data density.
>>
>>> Would
>>> they be fast enough to play HD videos and such? I have quite a few 1080
>>> HD videos. I don't want the drive to cause issues.
>> The transfer speed required for playing HD videos is virtually zero.
>> 1080p video compressed using an 8mbps rate require 2MB/s.  This can be
>> done even with the slowest drive from 10 years ago.  Today's slowest
>> drive are able to play about 40 or 50 of those HD video simultaneously.
>>    So the answer is yes.  They can play HD video :-)
>>
>> Most of those 5900/5400 disks are meant for pure data storage.  The
>> lower RPM is used to market them as "green and silent", meaning they
>> don't consume much power and aren't noisy.  Installing your OS on them
>> though isn't going to give you good speed.  They have good transfer
>> rates, but their access times usually suck.
>>
>>> Can someone that has one or more of these post their hdparm -Tt results?
>>> Different speeds would be great too. I'd like to compare what a 5400rpm
>>> drive would do compared to a 7200rpm drive.
>> Simply Google around for benchmarks of the drivers you're interested in.
>>    Note that is in area where it doesn't make any real difference that
>> the benches or reviews you find are performed under MS Windows.  The
>> results are applicable to every OS.
>>
>> As a rule of thumb when buying drives: if you want to install software
>> on it, buy an 7200RPM drive with good access times.  Of course they're
>> more expensive  If you just want to store all your downloaded HD porn
>> and music collection on it, a silent 5400RPM drive is a good choice.
>>
> indeed. Additionally they don't get really warm. Which reduces the overall
> thermal load in the case.
>
> One important thing:
>
> most if not all 2TB drives have 4K sectors, which means you have to be
> carefull while partitioning those beasts.


Looks like some good info.  I just need a GOOD sale and some extra money 
to spend.  Maybe in a couple weeks or so.  Hopefully. ;-)

As for heat in my case, I have a Cooler Master HAF-932 case.  It has 
those huge 230mm fans.  Heat is not a problem.

I just wonder how much data they will be able to pack into a 3.5" drive 
tho.  Hmmmmm.  Surely they will run out of room at some point.  I mean, 
the heads have got to have a little room to work with.

Dale

:-)  :-)

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Thu, Oct 27, 2011 at 1:30 PM, Dale <rdalek1967@gmail.com> wrote:
>
>
> Looks like some good info.  I just need a GOOD sale and some extra money to
> spend.  Maybe in a couple weeks or so.  Hopefully. ;-)
>
> As for heat in my case, I have a Cooler Master HAF-932 case.  It has those
> huge 230mm fans.  Heat is not a problem.
>
> I just wonder how much data they will be able to pack into a 3.5" drive tho.
>  Hmmmmm.  Surely they will run out of room at some point.  I mean, the heads
> have got to have a little room to work with.

Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
"Don't Buy Hard Drives Made By $x" experience.

Here's mine.

I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
sale a couple years ago. It failed on me. Around the same time, I
identified some flaws in the firmware which I considered severe[2].

I RMA'd the drive, including a full report on the failure and the bugs
I'd found in the firmware. I received the new drive in the mail. Same
exact model. Same exact firmware revision.[1] It failed on me within
three months. I attempted another RMA, the drive's serial number was
rejected by their system, and I never heard back.

So, I recommend not buying SAMSUNG drives for a combination of:
1) Historical evidence of poor firmware design. (reference smartctl's
man page; SAMSUNG is the only manufacturer I know of to get two
user-selectable workarounds in smartctl.)
2) I received a failed drive, which was RMA'd, the subsequent drive
failed shortly thereafter, and couldn't be RMA'd using normal
channels.
3) No acknowledgement (or even denial) of the firmware issue.

[1] Ok, sure, there's no way they'd be able to whip out a new firmware
revision in time for an RMA. That wouldn't make sense. But they might
have sent me a drive with a different firmware revision. Or a
different model. As it stood, they sent me back a device I'd already
identified as systemically defective.
[2] It claimed to support logging, but any failed test didn't get
appended to the log, but erased and replaced it. I can probably dig up
nearly all the details, but not quickly, since I'm at work. However,
since you're on the cusp of making a purchase, I thought I'd give you
fair warning...

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Dale]

Michael Mol wrote:
> On Thu, Oct 27, 2011 at 1:30 PM, Dale<rdalek1967@gmail.com>  wrote:
>>
>> Looks like some good info.  I just need a GOOD sale and some extra money to
>> spend.  Maybe in a couple weeks or so.  Hopefully. ;-)
>>
>> As for heat in my case, I have a Cooler Master HAF-932 case.  It has those
>> huge 230mm fans.  Heat is not a problem.
>>
>> I just wonder how much data they will be able to pack into a 3.5" drive tho.
>>   Hmmmmm.  Surely they will run out of room at some point.  I mean, the heads
>> have got to have a little room to work with.
> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
> "Don't Buy Hard Drives Made By $x" experience.
>
> Here's mine.
>
> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
> sale a couple years ago. It failed on me. Around the same time, I
> identified some flaws in the firmware which I considered severe[2].
>
> I RMA'd the drive, including a full report on the failure and the bugs
> I'd found in the firmware. I received the new drive in the mail. Same
> exact model. Same exact firmware revision.[1] It failed on me within
> three months. I attempted another RMA, the drive's serial number was
> rejected by their system, and I never heard back.
>
> So, I recommend not buying SAMSUNG drives for a combination of:
> 1) Historical evidence of poor firmware design. (reference smartctl's
> man page; SAMSUNG is the only manufacturer I know of to get two
> user-selectable workarounds in smartctl.)
> 2) I received a failed drive, which was RMA'd, the subsequent drive
> failed shortly thereafter, and couldn't be RMA'd using normal
> channels.
> 3) No acknowledgement (or even denial) of the firmware issue.
>
> [1] Ok, sure, there's no way they'd be able to whip out a new firmware
> revision in time for an RMA. That wouldn't make sense. But they might
> have sent me a drive with a different firmware revision. Or a
> different model. As it stood, they sent me back a device I'd already
> identified as systemically defective.
> [2] It claimed to support logging, but any failed test didn't get
> appended to the log, but erased and replaced it. I can probably dig up
> nearly all the details, but not quickly, since I'm at work. However,
> since you're on the cusp of making a purchase, I thought I'd give you
> fair warning...
>


To late now:

root@fireball / # hdparm -i /dev/sdc

/dev/sdc:

  Model=SAMSUNG HD753LJ, FwRev=1AA01117, SerialNo=S1PWJ1KS305193
  Config={ Fixed }
  RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
  BuffType=DualPortCache, BuffSize=unknown, MaxMultSect=16, MultSect=off
  CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=1465149168
  IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
  PIO modes:  pio0 pio1 pio2 pio3 pio4
  DMA modes:  mdma0 mdma1 mdma2
  UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
  AdvancedPM=yes: disabled (255) WriteCache=enabled
  Drive conforms to: unknown:  ATA/ATAPI-3,4,5,6,7

  * signifies the current active mode

root@fireball / #

I got this one about 2 years or so ago.  I did have random lockups a 
while back but I think it was a file system error.  I moved everything 
off the drive, reformatted it and it has worked fine ever since.  If I 
get me a new drive, the one above will be a backup sort of thing.

I seem to have good luck with WD and Maxtor myself.  Like you said tho, 
everyone has their horror story.  It is bad that they didn't give some 
sort of explanation on the second failure.  I have noticed that some 
things, car parts for example, have what they call a "limited 
warranty."  That means exchange once and then you are on your own if it 
fails.  Maybe they are doing that with their drives.  That would explain 
a lot too.

Dale

:-)  :-)

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Florian Philipp]

[-- Attachment #1: Type: text/plain, Size: 1272 bytes --]

Am 27.10.2011 19:30, schrieb Dale:
> Volker Armin Hemmann wrote:
>> Am Donnerstag 27 Oktober 2011, 13:09:17 schrieb Nikos Chantziaras:
>>> On 10/27/2011 11:15 AM, Dale wrote:
>>>> Howdy,
>>>>
>>>> I'm wanting to get a hard drive that is pretty good size. I'm looking
>>>> for about 1 to 2TBs or so. Thing is, a lot of them seem to be 5900 or
>>>> even 5400 rpm drives. I realize that the data on there is packed pretty
>>>> tight so I want to ask a few people that may have one or more of these
>>>> things a few questions. Are they as fast as a slower RPM drive?
>>> I assume you meant to say "as fast as a faster RPM drive".  No, of
>>> course not.  If we're speaking about the same capacity and amount of
>>> platters, of course.  If we're not, then yes, they can be as fast
>>> because of the higher data density.
>>>
[...]
>
> I just wonder how much data they will be able to pack into a 3.5" drive
> tho.  Hmmmmm.  Surely they will run out of room at some point.  I mean,
> the heads have got to have a little room to work with.
> 
> Dale
> 
> :-)  :-)
>

Well, then this story might cheer you up ;)
http://arstechnica.com/gadgets/news/2011/10/researchers-increase-hard-drive-density-sixfold-with-salt.ars

Regards,
Florian Philipp


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 262 bytes --]

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Donnerstag 27 Oktober 2011, 13:51:47 schrieb Michael Mol:
> On Thu, Oct 27, 2011 at 1:30 PM, Dale <rdalek1967@gmail.com> wrote:
> > Looks like some good info.  I just need a GOOD sale and some extra money
> > to spend.  Maybe in a couple weeks or so.  Hopefully. ;-)
> > 
> > As for heat in my case, I have a Cooler Master HAF-932 case.  It has
> > those huge 230mm fans.  Heat is not a problem.
> > 
> > I just wonder how much data they will be able to pack into a 3.5" drive
> > tho. Hmmmmm.  Surely they will run out of room at some point.  I mean,
> > the heads have got to have a little room to work with.
> 
> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
> "Don't Buy Hard Drives Made By $x" experience.
> 
> Here's mine.
> 
> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
> sale a couple years ago. It failed on me. Around the same time, I
> identified some flaws in the firmware which I considered severe[2].
> 
> I RMA'd the drive, including a full report on the failure and the bugs
> I'd found in the firmware. I received the new drive in the mail. Same
> exact model. Same exact firmware revision.[1] It failed on me within
> three months. I attempted another RMA, the drive's serial number was
> rejected by their system, and I never heard back.
> 
> So, I recommend not buying SAMSUNG drives for a combination of:
> 1) Historical evidence of poor firmware design. (reference smartctl's
> man page; SAMSUNG is the only manufacturer I know of to get two
> user-selectable workarounds in smartctl.)
> 2) I received a failed drive, which was RMA'd, the subsequent drive
> failed shortly thereafter, and couldn't be RMA'd using normal
> channels.
> 3) No acknowledgement (or even denial) of the firmware issue.
> 
> [1] Ok, sure, there's no way they'd be able to whip out a new firmware
> revision in time for an RMA. That wouldn't make sense. But they might
> have sent me a drive with a different firmware revision. Or a
> different model. As it stood, they sent me back a device I'd already
> identified as systemically defective.
> [2] It claimed to support logging, but any failed test didn't get
> appended to the log, but erased and replaced it. I can probably dig up
> nearly all the details, but not quickly, since I'm at work. However,
> since you're on the cusp of making a purchase, I thought I'd give you
> fair warning...

/dev/sda:

 Model=SAMSUNG MMCRE64G5MXP-0VB, FwRev=VBM1901Q, SerialNo=S0FDNEAZ600013
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=unknown, MaxMultSect=16, MultSect=16
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=125045424
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 
 AdvancedPM=no WriteCache=enabled
 Drive conforms to: ATA/ATAPI-7 T13 1532D revision 1:  ATA/ATAPI-2,3,4,5,6,7

 * signifies the current active mode


/dev/sdb:

 Model=SAMSUNG HD502IJ, FwRev=1AA01109, SerialNo=S13TJDWQ346413
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=16384kB, MaxMultSect=16, MultSect=off
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=976773168
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-3,4,5,6,7

 * signifies the current active mode


/dev/sdc:

 Model=SAMSUNG HD753LJ, FwRev=1AA01113, SerialNo=S13UJ1CQB07158
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=unknown, MaxMultSect=16, MultSect=off
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=1465149168
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-3,4,5,6,7

 * signifies the current active mode


/dev/sdd:

 Model=SAMSUNG HD502HJ, FwRev=1AJ100E5, SerialNo=S20BJDWS913888
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=off
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=976773168
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-0,1,2,3,4,5,6,7

 * signifies the current active mode


/dev/sde:

 Model=SAMSUNG HD103SJ, FwRev=1AJ10001, SerialNo=S246JD1Z910209
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=unknown, MaxMultSect=16, MultSect=off
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=1953525168
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-0,1,2,3,4,5,6,7

 * signifies the current active mode


/dev/sdf:

 Model=SAMSUNG HD103SJ, FwRev=1AJ100E5, SerialNo=S246JDWSA20722
 Config={ Fixed }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=unknown, MaxMultSect=16, MultSect=off
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=1953525168
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-0,1,2,3,4,5,6,7

 * signifies the current active mode

the 2tb drive is not connected at the moment - but, hey it's a Samsung - ans 
so quiet, that I sometimes forget to turn it off.

Oh, yeah it was THAT 2tb drive with the smart bug.

Which was solved with an easy to do firmware update.


-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Thu, Oct 27, 2011 at 2:52 PM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
> Am Donnerstag 27 Oktober 2011, 13:51:47 schrieb Michael Mol:
>> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
>> "Don't Buy Hard Drives Made By $x" experience.
>>
>> Here's mine.
>>
>> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
>> sale a couple years ago. It failed on me. Around the same time, I
>> identified some flaws in the firmware which I considered severe[2].
>>
>> I RMA'd the drive, including a full report on the failure and the bugs
>> I'd found in the firmware. I received the new drive in the mail. Same
>> exact model. Same exact firmware revision.[1] It failed on me within
>> three months. I attempted another RMA, the drive's serial number was
>> rejected by their system, and I never heard back.
>>
>> So, I recommend not buying SAMSUNG drives for a combination of:
>> 1) Historical evidence of poor firmware design. (reference smartctl's
>> man page; SAMSUNG is the only manufacturer I know of to get two
>> user-selectable workarounds in smartctl.)
>> 2) I received a failed drive, which was RMA'd, the subsequent drive
>> failed shortly thereafter, and couldn't be RMA'd using normal
>> channels.
>> 3) No acknowledgement (or even denial) of the firmware issue.
>>
>> [1] Ok, sure, there's no way they'd be able to whip out a new firmware
>> revision in time for an RMA. That wouldn't make sense. But they might
>> have sent me a drive with a different firmware revision. Or a
>> different model. As it stood, they sent me back a device I'd already
>> identified as systemically defective.
>> [2] It claimed to support logging, but any failed test didn't get
>> appended to the log, but erased and replaced it. I can probably dig up
>> nearly all the details, but not quickly, since I'm at work. However,
>> since you're on the cusp of making a purchase, I thought I'd give you
>> fair warning...
>
> /dev/sda:
>
>  Model=SAMSUNG MMCRE64G5MXP-0VB, FwRev=VBM1901Q, SerialNo=S0FDNEAZ600013
>
> /dev/sdb:
>
>  Model=SAMSUNG HD502IJ, FwRev=1AA01109, SerialNo=S13TJDWQ346413
>
> /dev/sdc:
>
>  Model=SAMSUNG HD753LJ, FwRev=1AA01113, SerialNo=S13UJ1CQB07158
>
> /dev/sdd:
>
>  Model=SAMSUNG HD502HJ, FwRev=1AJ100E5, SerialNo=S20BJDWS913888
>
> /dev/sde:
>
>  Model=SAMSUNG HD103SJ, FwRev=1AJ10001, SerialNo=S246JD1Z910209
>
> /dev/sdf:
>
>  Model=SAMSUNG HD103SJ, FwRev=1AJ100E5, SerialNo=S246JDWSA20722
>
> the 2tb drive is not connected at the moment - but, hey it's a Samsung - ans
> so quiet, that I sometimes forget to turn it off.
>
> Oh, yeah it was THAT 2tb drive with the smart bug.
>
> Which was solved with an easy to do firmware update.

I checked at the time. There was no firmware update, and, to my
knowledge, there never was for the drive model that failed on me.
Shortly after my second drive failed, Newegg discontinued selling
model. (The most I remember about the model number can be expressed as
a regex: HD10.*UI. I don't remember the firmware revision.

It was the combination of historical problems, personal incidental
experience and terrible customer service that led me to swear off
SAMSUNG drives. Take away any one of those issues from my experiences
at the time, and I'd consider buying another drive from them.

You've got six working drives of various sizes, models and firmware
revisions. Good for you. I've got a still-functional 40GB IBM
DeathStar. (It's not powered up right now, but it never failed on me
after five years of use.)

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Donnerstag 27 Oktober 2011, 15:17:45 schrieb Michael Mol:
> On Thu, Oct 27, 2011 at 2:52 PM, Volker Armin Hemmann
> 
> <volkerarmin@googlemail.com> wrote:
> > Am Donnerstag 27 Oktober 2011, 13:51:47 schrieb Michael Mol:
> >> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
> >> "Don't Buy Hard Drives Made By $x" experience.
> >> 
> >> Here's mine.
> >> 
> >> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
> >> sale a couple years ago. It failed on me. Around the same time, I
> >> identified some flaws in the firmware which I considered severe[2].
> >> 
> >> I RMA'd the drive, including a full report on the failure and the bugs
> >> I'd found in the firmware. I received the new drive in the mail. Same
> >> exact model. Same exact firmware revision.[1] It failed on me within
> >> three months. I attempted another RMA, the drive's serial number was
> >> rejected by their system, and I never heard back.
> >> 
> >> So, I recommend not buying SAMSUNG drives for a combination of:
> >> 1) Historical evidence of poor firmware design. (reference smartctl's
> >> man page; SAMSUNG is the only manufacturer I know of to get two
> >> user-selectable workarounds in smartctl.)
> >> 2) I received a failed drive, which was RMA'd, the subsequent drive
> >> failed shortly thereafter, and couldn't be RMA'd using normal
> >> channels.
> >> 3) No acknowledgement (or even denial) of the firmware issue.
> >> 
> >> [1] Ok, sure, there's no way they'd be able to whip out a new firmware
> >> revision in time for an RMA. That wouldn't make sense. But they might
> >> have sent me a drive with a different firmware revision. Or a
> >> different model. As it stood, they sent me back a device I'd already
> >> identified as systemically defective.
> >> [2] It claimed to support logging, but any failed test didn't get
> >> appended to the log, but erased and replaced it. I can probably dig up
> >> nearly all the details, but not quickly, since I'm at work. However,
> >> since you're on the cusp of making a purchase, I thought I'd give you
> >> fair warning...
> > 
> > /dev/sda:
> > 
> >  Model=SAMSUNG MMCRE64G5MXP-0VB, FwRev=VBM1901Q, SerialNo=S0FDNEAZ600013
> > 
> > /dev/sdb:
> > 
> >  Model=SAMSUNG HD502IJ, FwRev=1AA01109, SerialNo=S13TJDWQ346413
> > 
> > /dev/sdc:
> > 
> >  Model=SAMSUNG HD753LJ, FwRev=1AA01113, SerialNo=S13UJ1CQB07158
> > 
> > /dev/sdd:
> > 
> >  Model=SAMSUNG HD502HJ, FwRev=1AJ100E5, SerialNo=S20BJDWS913888
> > 
> > /dev/sde:
> > 
> >  Model=SAMSUNG HD103SJ, FwRev=1AJ10001, SerialNo=S246JD1Z910209
> > 
> > /dev/sdf:
> > 
> >  Model=SAMSUNG HD103SJ, FwRev=1AJ100E5, SerialNo=S246JDWSA20722
> > 
> > the 2tb drive is not connected at the moment - but, hey it's a Samsung -
> > ans so quiet, that I sometimes forget to turn it off.
> > 
> > Oh, yeah it was THAT 2tb drive with the smart bug.
> > 
> > Which was solved with an easy to do firmware update.
> 
> I checked at the time. There was no firmware update, and, to my
> knowledge, there never was for the drive model that failed on me.
> Shortly after my second drive failed, Newegg discontinued selling
> model. (The most I remember about the model number can be expressed as
> a regex: HD10.*UI. I don't remember the firmware revision.
> 
> It was the combination of historical problems, personal incidental
> experience and terrible customer service that led me to swear off
> SAMSUNG drives. Take away any one of those issues from my experiences
> at the time, and I'd consider buying another drive from them.
> 
> You've got six working drives of various sizes, models and firmware
> revisions. Good for you. I've got a still-functional 40GB IBM
> DeathStar. (It's not powered up right now, but it never failed on me
> after five years of use.)

and I had 5 death stars failing on me.

... 


-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Thu, Oct 27, 2011 at 3:41 PM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
> Am Donnerstag 27 Oktober 2011, 15:17:45 schrieb Michael Mol:
>> On Thu, Oct 27, 2011 at 2:52 PM, Volker Armin Hemmann
>>
>> <volkerarmin@googlemail.com> wrote:
>> > Am Donnerstag 27 Oktober 2011, 13:51:47 schrieb Michael Mol:
>> >> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
>> >> "Don't Buy Hard Drives Made By $x" experience.
>> >>
>> >> Here's mine.
>> >>
>> >> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black Friday
>> >> sale a couple years ago. It failed on me. Around the same time, I
>> >> identified some flaws in the firmware which I considered severe[2].
>> >  Model=SAMSUNG MMCRE64G5MXP-0VB, FwRev=VBM1901Q, SerialNo=S0FDNEAZ600013
>> >  Model=SAMSUNG HD502IJ, FwRev=1AA01109, SerialNo=S13TJDWQ346413
>> >  Model=SAMSUNG HD753LJ, FwRev=1AA01113, SerialNo=S13UJ1CQB07158
>> >  Model=SAMSUNG HD502HJ, FwRev=1AJ100E5, SerialNo=S20BJDWS913888
>> >  Model=SAMSUNG HD103SJ, FwRev=1AJ10001, SerialNo=S246JD1Z910209
>> >  Model=SAMSUNG HD103SJ, FwRev=1AJ100E5, SerialNo=S246JDWSA20722
>> >
>> > Oh, yeah it was THAT 2tb drive with the smart bug.
>> >
>> > Which was solved with an easy to do firmware update.
>>
>> I checked at the time. There was no firmware update, and, to my
>> knowledge, there never was for the drive model that failed on me.
>> Shortly after my second drive failed, Newegg discontinued selling
>> model. (The most I remember about the model number can be expressed as
>> a regex: HD10.*UI. I don't remember the firmware revision.
>>
>> It was the combination of historical problems, personal incidental
>> experience and terrible customer service that led me to swear off
>> SAMSUNG drives. Take away any one of those issues from my experiences
>> at the time, and I'd consider buying another drive from them.
>>
>> You've got six working drives of various sizes, models and firmware
>> revisions. Good for you. I've got a still-functional 40GB IBM
>> DeathStar. (It's not powered up right now, but it never failed on me
>> after five years of use.)
>
> and I had 5 death stars failing on me.

My point is that the numbers aren't what mattered here. My point is
that SAMSUNG sold me a shoddy product, replaced it with another
instance of the the same shoddy product, wouldn't replace it again,
and never addressed a detailed technical report of a systemic problem
in the same. Bad tech, bad customer service, and it looked like this
was a more common scenario than among other manufacturers. All of it
boiled down to a nasty case of being a bad candidate for spending time
and money.

Did IBM refuse to replace your failing drives? Did you include
detailed technical information that should have allowed them to
resolve issues leading to those drives' failures? For me, SAMSUNG's
behavior in the customer service department indicated that I wasn't
likely to get good service in the future, and the rapidly-failing
drives (combined with my analysis of the SMART output and the history
of SMART problems with SAMSUNG drives) indicated to me that I'd need
to use that customer service department in the future if I bought more
of their drives.

So you've got six working drives, and a drive that works now that you
patched the firmware. Congrats on choosing a model for which a
firmware patch was made available (unless that was just luck...).
Also, good luck if you have a failing drive that was sent to you by
RMA. It's been a few years; if you're lucky, they may have cleaned up
their act.

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Paul Hartman]

On Thu, Oct 27, 2011 at 12:51 PM, Michael Mol <mikemol@gmail.com> wrote:
> Just don't buy a SAMSUNG drive. I know, I know, everyone has their pet
> "Don't Buy Hard Drives Made By $x" experience.

On Thu, Oct 27, 2011 at 1:41 PM, Dale <rdalek1967@gmail.com> wrote:
> I seem to have good luck with WD and Maxtor myself.

Seagate and Western Digital have bought (or are in process of buying)
all of the other HDD manufacturers, except for Toshiba.

Seagate = Conner, Quantum, Maxtor, Samsung
Western Digital = Hitachi (= IBM)

Toshiba still stands alone, as far as I know, but they don't compete
in the consumer HDD market so, for your purposes, they don't exist
(unless you're building a large and expensive SAS array in your
house). :)

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Bill Longman]

On 10/27/2011 12:41 PM, Volker Armin Hemmann wrote:
> and I had 5 death stars failing on me.

Darth Vader's death star failed, too.

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Dale]

Florian Philipp wrote:
> Am 27.10.2011 19:30, schrieb Dale:
> I just wonder how much data they will be able to pack into a 3.5" drive
> tho.  Hmmmmm.  Surely they will run out of room at some point.  I mean,
> the heads have got to have a little room to work with.
>
> Dale
>
> :-)  :-)
>
> Well, then this story might cheer you up ;)
> http://arstechnica.com/gadgets/news/2011/10/researchers-increase-hard-drive-density-sixfold-with-salt.ars
>
> Regards,
> Florian Philipp
>

Good Golly Ms Molly.

"and have been able to fabricate magnetic storage media with a density 
of 3.3 terabits per square inch."

What is there about 10 or 12 square inches for a 3.5" drive?  That's 
about 30TBs.  O_O  That would take me a while to fill up even if I had a 
really fast DSL line.  WOW !!

Hmmmm, NCIS, all the CSI's, Criminal Minds, Numb3rs, and lots of 
others.  Heck, I could cut off DirecTv then.  Just get that Hulu thingy 
and go nuts.  Does Hulu work with Linux?  Seems I read it doesn't.  
Anyway, still a LOT of shows.

Dale

:-)  :-)

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Mark Knecht]

On Thu, Oct 27, 2011 at 9:10 PM, Dale <rdalek1967@gmail.com> wrote:
> Florian Philipp wrote:
>>
>> Am 27.10.2011 19:30, schrieb Dale:
>> I just wonder how much data they will be able to pack into a 3.5" drive
>> tho.  Hmmmmm.  Surely they will run out of room at some point.  I mean,
>> the heads have got to have a little room to work with.
>>
>> Dale
>>
>> :-)  :-)
>>
>> Well, then this story might cheer you up ;)
>>
>> http://arstechnica.com/gadgets/news/2011/10/researchers-increase-hard-drive-density-sixfold-with-salt.ars
>>
>> Regards,
>> Florian Philipp
>>
>
> Good Golly Ms Molly.
>
> "and have been able to fabricate magnetic storage media with a density of
> 3.3 terabits per square inch."
>
> What is there about 10 or 12 square inches for a 3.5" drive?  That's about
> 30TBs.  O_O  That would take me a while to fill up even if I had a really
> fast DSL line.  WOW !!
>
> Hmmmm, NCIS, all the CSI's, Criminal Minds, Numb3rs, and lots of others.
>  Heck, I could cut off DirecTv then.  Just get that Hulu thingy and go nuts.
>  Does Hulu work with Linux?  Seems I read it doesn't.  Anyway, still a LOT
> of shows.
>
> Dale

Hulu works with Linux last time I tried but I typically use a Windows
VM to watch it..

NetFlix works with Linux but only in a Windows VM.

Amazon streams now but I haven't tried them yet. (Watch out for 'One
Touch' and accidentally spending money...)

Most of the TV networks have web site streaming of their programs
which work in Linux.

I vaguely remember reading somewhere that it would take 6TB to
digitize the Library of Congress. Whether that's true or not, we're
close to the point where every person can have a personal copy of the
Library of Congress.

- Mark

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Dale]

Mark Knecht wrote:
> On Thu, Oct 27, 2011 at 9:10 PM, Dale<rdalek1967@gmail.com>  wrote:
>> Florian Philipp wrote:
>>> Am 27.10.2011 19:30, schrieb Dale:
>>> I just wonder how much data they will be able to pack into a 3.5" drive
>>> tho.  Hmmmmm.  Surely they will run out of room at some point.  I mean,
>>> the heads have got to have a little room to work with.
>>>
>>> Dale
>>>
>>> :-)  :-)
>>>
>>> Well, then this story might cheer you up ;)
>>>
>>> http://arstechnica.com/gadgets/news/2011/10/researchers-increase-hard-drive-density-sixfold-with-salt.ars
>>>
>>> Regards,
>>> Florian Philipp
>>>
>> Good Golly Ms Molly.
>>
>> "and have been able to fabricate magnetic storage media with a density of
>> 3.3 terabits per square inch."
>>
>> What is there about 10 or 12 square inches for a 3.5" drive?  That's about
>> 30TBs.  O_O  That would take me a while to fill up even if I had a really
>> fast DSL line.  WOW !!
>>
>> Hmmmm, NCIS, all the CSI's, Criminal Minds, Numb3rs, and lots of others.
>>   Heck, I could cut off DirecTv then.  Just get that Hulu thingy and go nuts.
>>   Does Hulu work with Linux?  Seems I read it doesn't.  Anyway, still a LOT
>> of shows.
>>
>> Dale
> Hulu works with Linux last time I tried but I typically use a Windows
> VM to watch it..
>
> NetFlix works with Linux but only in a Windows VM.
>
> Amazon streams now but I haven't tried them yet. (Watch out for 'One
> Touch' and accidentally spending money...)
>
> Most of the TV networks have web site streaming of their programs
> which work in Linux.
>
> I vaguely remember reading somewhere that it would take 6TB to
> digitize the Library of Congress. Whether that's true or not, we're
> close to the point where every person can have a personal copy of the
> Library of Congress.
>
> - Mark
>
>


I have to say this.  I worked at a computer place in the late 80's.  
Back then a 25Mhz CPU was fast as lightening.  Hard drives were maybe 
100MBs.  Memory, lucky if you have a 1 or 2Mbs.  Now look where we are.  
I'm still trying to play Steve Jobs and imagine where we will be 20 
years from now.

I'm real bad to download videos.  I have been working on NCIS here the 
past few weeks.  I try to get HD when I can.  One day I hope to hook my 
TV up to my puter.  I have a HDMI connector on the video card and my TV 
has a couple on the back of it.  I just don't know yet how to make them 
work or if they work already.  I need to test that one day.  Anyway, if 
I get enough stuff downloaded, I could in theory suspend DirecTv for a 
month or two.  Put the money on some bills.

Oh, I use downloadhelper to get the videos.  It stopped working with 
Seamonkey 2.4 so back to using Firefox right now.  It keeps crashing 
tho.  Still scratching my head on that one.  ;-)

Hulu works with Linux huh?  I got to remember that.  I did know that 
Netflix didn't tho.  I thought about trying it but they went up.

Life.

Dale

:-)  :-)

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Fri, Oct 28, 2011 at 7:36 AM, Dale <rdalek1967@gmail.com> wrote:
> Mark Knecht wrote:
>>
> I have to say this.  I worked at a computer place in the late 80's.  Back
> then a 25Mhz CPU was fast as lightening.  Hard drives were maybe 100MBs.
>  Memory, lucky if you have a 1 or 2Mbs.  Now look where we are.  I'm still
> trying to play Steve Jobs and imagine where we will be 20 years from now.

I've got a 157MB hard drive at home. My bio-dad gave it to me earlier
this week in a stack of hard drives he was getting rid of.

As for 80s computers...our first home computer was a Tandy RLX 1000. I
convinced my mother to spend our savings on it, as I'd gotten hooked
on programming with an Apple ][ at school. Kinda proud of her; she
picked up programming quickly, took a few classes, really took to
databases, and now she's a manager in the IT department at the local
community college.

>
> I'm real bad to download videos.  I have been working on NCIS here the past
> few weeks.  I try to get HD when I can.  One day I hope to hook my TV up to
> my puter.  I have a HDMI connector on the video card and my TV has a couple
> on the back of it.  I just don't know yet how to make them work or if they
> work already.  I need to test that one day.

It's really, really easy if you've ever done multimon under Linux.  A
television on that HDMI port will behave pretty much like a DVI
monitor.

>  Anyway, if I get enough stuff
> downloaded, I could in theory suspend DirecTv for a month or two.  Put the
> money on some bills.
>
> Oh, I use downloadhelper to get the videos.  It stopped working with
> Seamonkey 2.4 so back to using Firefox right now.  It keeps crashing tho.
>  Still scratching my head on that one.  ;-)
>
> Hulu works with Linux huh?  I got to remember that.  I did know that Netflix
> didn't tho.  I thought about trying it but they went up.

Hulu works, Netflix doesn't. Amazon may, but if you have a network
glitch and their server doesn't realize you're not pulling data any
more, you have to wait a while before you try again. At least, that
was my experience with them a bit over a year ago. (Real annoying when
you go to show something to a group of friends on a/v night. Kills the
mood and flow.)

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Paul Hartman]

On Fri, Oct 28, 2011 at 6:15 AM, Mark Knecht <markknecht@gmail.com> wrote:
> Amazon streams now but I haven't tried them yet. (Watch out for 'One
> Touch' and accidentally spending money...)

Amazon streaming works fine in Linux (I've tried it), it just uses the
plain old Adobe Flash plug-in.

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Donnerstag 27 Oktober 2011, 16:00:09 schrieb Michael Mol:
> On Thu, Oct 27, 2011 at 3:41 PM, Volker Armin Hemmann
> 
> <volkerarmin@googlemail.com> wrote:
> > Am Donnerstag 27 Oktober 2011, 15:17:45 schrieb Michael Mol:
> >> On Thu, Oct 27, 2011 at 2:52 PM, Volker Armin Hemmann
> >> 
> >> <volkerarmin@googlemail.com> wrote:
> >> > Am Donnerstag 27 Oktober 2011, 13:51:47 schrieb Michael Mol:
> >> >> Just don't buy a SAMSUNG drive. I know, I know, everyone has
> >> >> their pet "Don't Buy Hard Drives Made By $x" experience.
> >> >> 
> >> >> Here's mine.
> >> >> 
> >> >> I bought a 1TB SAMSUNG drive for cheap from Newegg at a Black
> >> >> Friday
> >> >> sale a couple years ago. It failed on me. Around the same time,
> >> >> I
> >> >> identified some flaws in the firmware which I considered
> >> >> severe[2].
> >> > 
> >> >  Model=SAMSUNG MMCRE64G5MXP-0VB, FwRev=VBM1901Q,
> >> > SerialNo=S0FDNEAZ600013 Model=SAMSUNG HD502IJ, FwRev=1AA01109,
> >> > SerialNo=S13TJDWQ346413 Model=SAMSUNG HD753LJ, FwRev=1AA01113,
> >> > SerialNo=S13UJ1CQB07158 Model=SAMSUNG HD502HJ, FwRev=1AJ100E5,
> >> > SerialNo=S20BJDWS913888 Model=SAMSUNG HD103SJ, FwRev=1AJ10001,
> >> > SerialNo=S246JD1Z910209 Model=SAMSUNG HD103SJ, FwRev=1AJ100E5,
> >> > SerialNo=S246JDWSA20722
> >> > 
> >> > Oh, yeah it was THAT 2tb drive with the smart bug.
> >> > 
> >> > Which was solved with an easy to do firmware update.
> >> 
> >> I checked at the time. There was no firmware update, and, to my
> >> knowledge, there never was for the drive model that failed on me.
> >> Shortly after my second drive failed, Newegg discontinued selling
> >> model. (The most I remember about the model number can be expressed as
> >> a regex: HD10.*UI. I don't remember the firmware revision.
> >> 
> >> It was the combination of historical problems, personal incidental
> >> experience and terrible customer service that led me to swear off
> >> SAMSUNG drives. Take away any one of those issues from my experiences
> >> at the time, and I'd consider buying another drive from them.
> >> 
> >> You've got six working drives of various sizes, models and firmware
> >> revisions. Good for you. I've got a still-functional 40GB IBM
> >> DeathStar. (It's not powered up right now, but it never failed on me
> >> after five years of use.)
> > 
> > and I had 5 death stars failing on me.
> 
> My point is that the numbers aren't what mattered here. My point is
> that SAMSUNG sold me a shoddy product, replaced it with another
> instance of the the same shoddy product, wouldn't replace it again,
> and never addressed a detailed technical report of a systemic problem
> in the same. Bad tech, bad customer service, and it looked like this
> was a more common scenario than among other manufacturers. All of it
> boiled down to a nasty case of being a bad candidate for spending time
> and money.

Samsung replaced my two drives (one 500gb drive from a known shady series and 
the SSD when the firmware overwrote itself) without any fuss.

Of course, I included the results of their check tool. No problem.

> 
> Did IBM refuse to replace your failing drives? 

no, but 5 out of 5 died and took tons of valuable data with them.

That is the worst case scenario. All other harddrives had no problems at all. 
Different mobos, PSUs - nothing changed the fact that all death stars I ever 
owned died violently - except the last one, because I sold the computer it was 
built in in time.

> Did you include
> detailed technical information that should have allowed them to
> resolve issues leading to those drives' failures? For me, SAMSUNG's
> behavior in the customer service department indicated that I wasn't
> likely to get good service in the future, and the rapidly-failing
> drives (combined with my analysis of the SMART output and the history
> of SMART problems with SAMSUNG drives) indicated to me that I'd need
> to use that customer service department in the future if I bought more
> of their drives.

there was one smart related problem with Samsung in the last year. With their 
2tb drives. Samsung released a firmware patch after they were informed of the 
problem.

Apart from that Samsung drives just work for me - and the people around me.

I also had no problems with drives getting replaced - but the replacing was 
always the problem of my trusted hardware dealer. That is why you buy hdds 
from a trusted, local guy. The rest was the problem of Samsung. They can't 
wiggle out of Germany's warranty laws ;)

> 
> So you've got six working drives, and a drive that works now that you
> patched the firmware. Congrats on choosing a model for which a
> firmware patch was made available (unless that was just luck...).
> Also, good luck if you have a failing drive that was sent to you by
> RMA. It's been a few years; if you're lucky, they may have cleaned up
> their act.

In fact there were two: a 500gb from a known series, and the SSD. Both got 
replaced because both failed in the warranty period. They had no chance but 
had to replace them. It went quickly and without any fuss. I did run their 
check tool and reported the results. No more questions asked. Here is the old 
one, thanks for the new one.

Western Digital - two drives. They worked. They were lousy, noisy, lame, hot, 
but worked.

Seagate: now 3 drives, 1 had to be replaced. It was pretty much dead on 
arrival, with a severe spindle defect. All three worked just fine until I 
replaced them

Toshiba&co have to check my Dell Poweredge what drives are in there...

IBM all death stars I ever owned failed except the last one - which was given 
away (and failed with its new owner). Which is the worst possible outcome. No 
customer support can outweight such a disaster.


-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

(oi, it's getting harder and harder to snip this thing properly)

On Fri, Oct 28, 2011 at 10:55 AM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
> Am Donnerstag 27 Oktober 2011, 16:00:09 schrieb Michael Mol:
>> On Thu, Oct 27, 2011 at 3:41 PM, Volker Armin Hemmann
>> <volkerarmin@googlemail.com> wrote:
>> > Am Donnerstag 27 Oktober 2011, 15:17:45 schrieb Michael Mol:
>> >> You've got six working drives of various sizes, models and firmware
>> >> revisions. Good for you. I've got a still-functional 40GB IBM
>> >> DeathStar. (It's not powered up right now, but it never failed on me
>> >> after five years of use.)
>> >
>> > and I had 5 death stars failing on me.
>>
>> My point is that the numbers aren't what mattered here. My point is
>> that SAMSUNG sold me a shoddy product, replaced it with another
>> instance of the the same shoddy product, wouldn't replace it again,
>> and never addressed a detailed technical report of a systemic problem
>> in the same. Bad tech, bad customer service, and it looked like this
>> was a more common scenario than among other manufacturers. All of it
>> boiled down to a nasty case of being a bad candidate for spending time
>> and money.
>
> Samsung replaced my two drives (one 500gb drive from a known shady series and
> the SSD when the firmware overwrote itself) without any fuss.
>
> Of course, I included the results of their check tool. No problem.
>
>>
>> Did IBM refuse to replace your failing drives?
>
> no, but 5 out of 5 died and took tons of valuable data with them.
>
> That is the worst case scenario. All other harddrives had no problems at all.
> Different mobos, PSUs - nothing changed the fact that all death stars I ever
> owned died violently - except the last one, because I sold the computer it was
> built in in time.
>
>> Did you include
>> detailed technical information that should have allowed them to
>> resolve issues leading to those drives' failures? For me, SAMSUNG's
>> behavior in the customer service department indicated that I wasn't
>> likely to get good service in the future, and the rapidly-failing
>> drives (combined with my analysis of the SMART output and the history
>> of SMART problems with SAMSUNG drives) indicated to me that I'd need
>> to use that customer service department in the future if I bought more
>> of their drives.
>
> there was one smart related problem with Samsung in the last year. With their
> 2tb drives. Samsung released a firmware patch after they were informed of the
> problem.
>
> Apart from that Samsung drives just work for me - and the people around me.
>
> I also had no problems with drives getting replaced - but the replacing was
> always the problem of my trusted hardware dealer. That is why you buy hdds
> from a trusted, local guy. The rest was the problem of Samsung. They can't
> wiggle out of Germany's warranty laws ;)

Yeah, I bought mine from Newegg. Hard drive warranty through Newegg
means "see manufacturer". I've had no problem with WD's customer
service. Haven't had occasion to try anyone but SAMSUNG and WD.

Problem with buying locally where I am is that it generally means a
50% markup above what I'd pay Newegg. There are several big-box stores
in the area whose prices and selections aren't great. There are three
small computer stores within a 45 minute drive, and their selection
tends to be better, but their prices tend to be worse.

>> So you've got six working drives, and a drive that works now that you
>> patched the firmware. Congrats on choosing a model for which a
>> firmware patch was made available (unless that was just luck...).
>> Also, good luck if you have a failing drive that was sent to you by
>> RMA. It's been a few years; if you're lucky, they may have cleaned up
>> their act.
>
> In fact there were two: a 500gb from a known series, and the SSD. Both got
> replaced because both failed in the warranty period. They had no chance but
> had to replace them. It went quickly and without any fuss. I did run their
> check tool and reported the results. No more questions asked. Here is the old
> one, thanks for the new one.

Heh.

> Western Digital - two drives. They worked. They were lousy, noisy, lame, hot,
> but worked.

Yup. I've had a couple WD drives fail, but most were replaced for
other reasons first. I've got three WD drives spinning at home in
various machines.

> Seagate: now 3 drives, 1 had to be replaced. It was pretty much dead on
> arrival, with a severe spindle defect. All three worked just fine until I
> replaced them

Loving my Seagates. I've got four Seagates spinning at home, including
one that's been my core drive since 2005 or 2005.

> Toshiba&co have to check my Dell Poweredge what drives are in there...

Yeah, I don't have any current Toshibas.

> IBM all death stars I ever owned failed except the last one - which was given
> away (and failed with its new owner). Which is the worst possible outcome. No
> customer support can outweight such a disaster.

Not a unique story, by any stretch. IBM earned a reputation on that
line. My point with bringing up my DeathStar was to highlight that all
manufacturers have their good and bad drives; I had 2/2 bad SAMSUNG
drives, you've have 6/8 good ones. I've had 1/1 good DeathStar drives,
you've had 6/6 bad ones.

My chief complaint with SAMSUNG was their failure rate and terrible
customer service followup, at least per my experience. I notified them
of the firmware flaw, they ignored it, and it bit me again when the
replacement drive failed. Even if my experience _were_ a contemporary
statistical fluke, your account suggests their current state doesn't
match my experience, at least in terms of failure rate. You never
dealt with their customer service for replacements, so there are
obviously workarounds to that side of my experience if one is willing
to pay a bit more.

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Mark Knecht]

On Fri, Oct 28, 2011 at 7:18 AM, Paul Hartman
<paul.hartman+gentoo@gmail.com> wrote:
> On Fri, Oct 28, 2011 at 6:15 AM, Mark Knecht <markknecht@gmail.com> wrote:
>> Amazon streams now but I haven't tried them yet. (Watch out for 'One
>> Touch' and accidentally spending money...)
>
> Amazon streaming works fine in Linux (I've tried it), it just uses the
> plain old Adobe Flash plug-in.
>
>

Therein apparently lies the problem for me. Adobe Flash no longer
works on my machine since adding a second video card. I just tried
Amazon in Firefox 7 and it just sits doing nothing.

I _think_ this is related to using Xinerama on KDE but when I posted
questions here (twice) I got no responses so it seems few people are
doing this. For my futures trading I need 3 or 4 monitors and I
couldn't make sense out of using X without Xinerama so I seem stuck,
at least native in Gentoo. I suspect it's fine in a Windows VM but
I've got Jimmy Kimmel running at the moment so I'll test that later.

I REALLY, REALLY miss having all the OpenGL stuff you get with KDE. I
had it with one Nvidia card but not with two...

- Mark

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Freitag 28 Oktober 2011, 11:19:37 schrieb Michael Mol:
> (oi, it's getting harder and harder to snip this thing properly)
> 
> On Fri, Oct 28, 2011 at 10:55 AM, Volker Armin Hemmann
> 
> <volkerarmin@googlemail.com> wrote:
> > Am Donnerstag 27 Oktober 2011, 16:00:09 schrieb Michael Mol:
> >> On Thu, Oct 27, 2011 at 3:41 PM, Volker Armin Hemmann
> >> 
> >> <volkerarmin@googlemail.com> wrote:
> >> > Am Donnerstag 27 Oktober 2011, 15:17:45 schrieb Michael Mol:
> >> >> You've got six working drives of various sizes, models and
> >> >> firmware
> >> >> revisions. Good for you. I've got a still-functional 40GB IBM
> >> >> DeathStar. (It's not powered up right now, but it never failed
> >> >> on me
> >> >> after five years of use.)
> >> > 
> >> > and I had 5 death stars failing on me.
> >> 
> >> My point is that the numbers aren't what mattered here. My point is
> >> that SAMSUNG sold me a shoddy product, replaced it with another
> >> instance of the the same shoddy product, wouldn't replace it again,
> >> and never addressed a detailed technical report of a systemic problem
> >> in the same. Bad tech, bad customer service, and it looked like this
> >> was a more common scenario than among other manufacturers. All of it
> >> boiled down to a nasty case of being a bad candidate for spending time
> >> and money.
> > 
> > Samsung replaced my two drives (one 500gb drive from a known shady
> > series and the SSD when the firmware overwrote itself) without any
> > fuss.
> > 
> > Of course, I included the results of their check tool. No problem.
> > 
> >> Did IBM refuse to replace your failing drives?
> > 
> > no, but 5 out of 5 died and took tons of valuable data with them.
> > 
> > That is the worst case scenario. All other harddrives had no problems at
> > all. Different mobos, PSUs - nothing changed the fact that all death
> > stars I ever owned died violently - except the last one, because I sold
> > the computer it was built in in time.
> > 
> >> Did you include
> >> detailed technical information that should have allowed them to
> >> resolve issues leading to those drives' failures? For me, SAMSUNG's
> >> behavior in the customer service department indicated that I wasn't
> >> likely to get good service in the future, and the rapidly-failing
> >> drives (combined with my analysis of the SMART output and the history
> >> of SMART problems with SAMSUNG drives) indicated to me that I'd need
> >> to use that customer service department in the future if I bought more
> >> of their drives.
> > 
> > there was one smart related problem with Samsung in the last year. With
> > their 2tb drives. Samsung released a firmware patch after they were
> > informed of the problem.
> > 
> > Apart from that Samsung drives just work for me - and the people around
> > me.
> > 
> > I also had no problems with drives getting replaced - but the replacing
> > was always the problem of my trusted hardware dealer. That is why you
> > buy hdds from a trusted, local guy. The rest was the problem of
> > Samsung. They can't wiggle out of Germany's warranty laws ;)
> 
> Yeah, I bought mine from Newegg. Hard drive warranty through Newegg
> means "see manufacturer". I've had no problem with WD's customer
> service. Haven't had occasion to try anyone but SAMSUNG and WD.
> 
> Problem with buying locally where I am is that it generally means a
> 50% markup above what I'd pay Newegg. There are several big-box stores
> in the area whose prices and selections aren't great. There are three
> small computer stores within a 45 minute drive, and their selection
> tends to be better, but their prices tend to be worse.

my 'local' shop has only 10% higher prices than cheap online sellers. Worst 
case. Most of the time it is a lot less - and sometimes cheaper than online. I 
am willing to pay that 'premium' for service and the knowledge that when I 
tell them the drive is dead they don't even try to put the blame on me. Same 
for Ram - or mainboards ;) After bad experience with Enermax - waiting 6 weeks 
for a replacement for a PSU that killed three mobos and enermax being too 
cheap to even put a new set of cables in the box I went with BeQuiet. 48h 
replacement.

-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Freitag 28 Oktober 2011, 08:53:40 schrieb Mark Knecht:
> On Fri, Oct 28, 2011 at 7:18 AM, Paul Hartman
> 
> <paul.hartman+gentoo@gmail.com> wrote:
> > On Fri, Oct 28, 2011 at 6:15 AM, Mark Knecht <markknecht@gmail.com> wrote:
> >> Amazon streams now but I haven't tried them yet. (Watch out for 'One
> >> Touch' and accidentally spending money...)
> > 
> > Amazon streaming works fine in Linux (I've tried it), it just uses the
> > plain old Adobe Flash plug-in.
> 
> Therein apparently lies the problem for me. Adobe Flash no longer
> works on my machine since adding a second video card. I just tried
> Amazon in Firefox 7 and it just sits doing nothing.
> 
> I _think_ this is related to using Xinerama on KDE but when I posted
> questions here (twice) I got no responses so it seems few people are
> doing this. For my futures trading I need 3 or 4 monitors and I
> couldn't make sense out of using X without Xinerama so I seem stuck,
> at least native in Gentoo. I suspect it's fine in a Windows VM but
> I've got Jimmy Kimmel running at the moment so I'll test that later.
> 
> I REALLY, REALLY miss having all the OpenGL stuff you get with KDE. I
> had it with one Nvidia card but not with two...
> 
> - Mark

you don't need xinerama for multi-monitor setups.

-- 
#163933

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Mark Knecht]

On Fri, Oct 28, 2011 at 12:11 PM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
> Am Freitag 28 Oktober 2011, 08:53:40 schrieb Mark Knecht:
>> On Fri, Oct 28, 2011 at 7:18 AM, Paul Hartman
>>
>> <paul.hartman+gentoo@gmail.com> wrote:
>> > On Fri, Oct 28, 2011 at 6:15 AM, Mark Knecht <markknecht@gmail.com> wrote:
>> >> Amazon streams now but I haven't tried them yet. (Watch out for 'One
>> >> Touch' and accidentally spending money...)
>> >
>> > Amazon streaming works fine in Linux (I've tried it), it just uses the
>> > plain old Adobe Flash plug-in.
>>
>> Therein apparently lies the problem for me. Adobe Flash no longer
>> works on my machine since adding a second video card. I just tried
>> Amazon in Firefox 7 and it just sits doing nothing.
>>
>> I _think_ this is related to using Xinerama on KDE but when I posted
>> questions here (twice) I got no responses so it seems few people are
>> doing this. For my futures trading I need 3 or 4 monitors and I
>> couldn't make sense out of using X without Xinerama so I seem stuck,
>> at least native in Gentoo. I suspect it's fine in a Windows VM but
>> I've got Jimmy Kimmel running at the moment so I'll test that later.
>>
>> I REALLY, REALLY miss having all the OpenGL stuff you get with KDE. I
>> had it with one Nvidia card but not with two...
>>
>> - Mark
>
> you don't need xinerama for multi-monitor setups.
>

Volker,
   I understand that but I'm wondering if you read my post? I stated
'I couldn't make sense out of using X without Xinerama'.

   If you're interested in discussing multiple graphic card,
multi-monitor setups then I'd be more than willing to start another
thread. However that discussion doesn't belong in a thread entitled
"Hard drive RPMs and data speed".

- Mark

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[daid kahl]

Lots of HDD RPMs and company suggestions, but to the point...

My two cents are: ESATA.  I have multi TB external disks which I have
physics data stored on and needs to be analyzed.  USB might be as fast
(in the best case), but it uses processor overhead.

Not that a lot of machines support that kind of input.  I picked up a
decent laptop for cheap that also supports ESATA.  I didn't do
benchmarks or anything, but it's really insane IMO.

~daid

PS Sorry I deleted all the reply text.  I didn't want to copy/paste
individual references to different company external drives and so on,
just to not really care.  Mine is something by Buffalo, but I care
because it has ESATA.

PPS Or you could be my friends using USB formated NTFS and I can use
top to see how much processor power is used by ntfs-3g just to read
the data.  Ugh!

Re: [gentoo-user] Hard drive RPMs and data speed.

[James Broadhead]

On 27 October 2011 09:15, Dale <rdalek1967@gmail.com> wrote:
> Howdy,
>
> I'm wanting to get a hard drive that is pretty good size.  I'm looking for
> about 1 to 2TBs or so.  Thing is, a lot of them seem to be 5900 or even 5400
> rpm drives.  I realize that the data on there is packed pretty tight so I
> want to ask a few people that may have one or more of these things a few
> questions.  Are they as fast as a slower RPM drive?  Would they be fast
> enough to play HD videos and such?  I have quite a few 1080 HD videos.  I
> don't want the drive to cause issues.

Ignoring your question somewhat, since the hdparm test won't actually
get you 'effective' throughput, only 'ideal'.

( (4.4*1024*1024*1024) / (120*60) ) / 1024
640.796

So a 4.4GiB movie that lasts 2 hours would require a sustained drive
throughput of 640KiB/s  - which is pretty achievable.

My experience says that it doesn't matter how slow a drive you use,
provided that you beef up mplayer's cache size and minimum cache
threshold, since my laptop has a slow drive that likes to power down,
but loads of RAM.

>> grep cache /etc/mplayer/mplayer.conf
# cache settings
# Use 8MB input cache by default.
cache = 131072
# Prefill 20% of the cache before starting playback.
cache-min = 20.0
# Prefill 50% of the cache before restarting playback after the cache emptied.
cache-seek-min = 50

Re: [gentoo-user] Hard drive RPMs and data speed.

[Dale]

James Broadhead wrote:
> On 27 October 2011 09:15, Dale<rdalek1967@gmail.com>  wrote:
>> Howdy,
>>
>> I'm wanting to get a hard drive that is pretty good size.  I'm looking for
>> about 1 to 2TBs or so.  Thing is, a lot of them seem to be 5900 or even 5400
>> rpm drives.  I realize that the data on there is packed pretty tight so I
>> want to ask a few people that may have one or more of these things a few
>> questions.  Are they as fast as a slower RPM drive?  Would they be fast
>> enough to play HD videos and such?  I have quite a few 1080 HD videos.  I
>> don't want the drive to cause issues.
> Ignoring your question somewhat, since the hdparm test won't actually
> get you 'effective' throughput, only 'ideal'.
>
> ( (4.4*1024*1024*1024) / (120*60) ) / 1024
> 640.796
>
> So a 4.4GiB movie that lasts 2 hours would require a sustained drive
> throughput of 640KiB/s  - which is pretty achievable.
>
> My experience says that it doesn't matter how slow a drive you use,
> provided that you beef up mplayer's cache size and minimum cache
> threshold, since my laptop has a slow drive that likes to power down,
> but loads of RAM.
>
>>> grep cache /etc/mplayer/mplayer.conf
> # cache settings
> # Use 8MB input cache by default.
> cache = 131072
> # Prefill 20% of the cache before starting playback.
> cache-min = 20.0
> # Prefill 50% of the cache before restarting playback after the cache emptied.
> cache-seek-min = 50
>
>

I have the same here too.  Like minds maybe?  o_O

Dale

:-)  :-)

Re: [gentoo-user] Hard drive RPMs and data speed.

[James Broadhead]

On 31 October 2011 10:58, Dale <rdalek1967@gmail.com> wrote:
> James Broadhead wrote:
>> # Use 8MB input cache by default.
>> cache = 131072
>
> I have the same here too.  Like minds maybe?  o_O

I _think_ that it's the highest power-of-two that mplayer will allow
... so maybe not.

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Alex Schuster]

Michael Mol writes:

> My point is that the numbers aren't what mattered here. My point is
> that SAMSUNG sold me a shoddy product, replaced it with another
> instance of the the same shoddy product, wouldn't replace it again,
> and never addressed a detailed technical report of a systemic problem
> in the same. Bad tech, bad customer service, and it looked like this
> was a more common scenario than among other manufacturers. All of it
> boiled down to a nasty case of being a bad candidate for spending time
> and money.

Samsung, uh? Here's my story of today. My fried just bought two external
USB drives. I wanted to know which brand the HD is, so I checked with
hdparm -I, and googled for SAMSUNG HD204UI. I found a story about a bug
which makes the drive sometimes forget to write a block when it is
attached to a SATA adapter in AHCI mode and when the ATA command
"IDENTIFY DEVICE" is sent (like in hdparm -I or when using the
smartmontools). There is a firmware patch for this, this is good. But on
the annoying side:

- You need to make a DOS boot floppy and copy the patch there. I don't
  know how exactly to do this, and I read about people using Linux who
  needed over an hour for this or even failed. Can't they just let me
  download an image I can boot from?

- It doesn't work over USB, so I would have to install the drive in a PC.

- The new firmware has exactly the same revision number. How stupid is
  this?? I cannot even find out whether the drives have the problem or
  not. Except by trying to reproduce the problem.

Here's a link to the but I described, but It's German only.
http://www.heise.de/ct/meldung/Firmware-Patch-fuer-Samsung-Festplatte-EcoGreen-F4-HD204UI-Update-1150154.html
I also read some angry comments about Samsung there. Question is, are
other manufacturers better? And wasn't Samsung Electronics bought by
Seagate anyway?


Any idea whether an external USB drive case might count as a SATA
controller in AHCI mode? I tried to trigger the bug, but that did not
happen, so I guess it's fine, at least when being in the USB case.

Another problem is that data access frequently stalls on her PC, like when
transferring data or doing a mke2fs. After a while, this message appears
in syslog, and the process continues for a while, until it happens again:

usb 1-4: reset high speed USB device using ehci_hcd and address 7

Same problem with a GRML boot cd and on another USB port. Happens with
both drives. But it is fine on my PC. 

	Wonko

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Michael Mol]

On Mon, Oct 31, 2011 at 12:21 PM, Alex Schuster <wonko@wonkology.org> wrote:
> Michael Mol writes:
>
>> My point is that the numbers aren't what mattered here. My point is
>> that SAMSUNG sold me a shoddy product, replaced it with another
>> instance of the the same shoddy product, wouldn't replace it again,
>> and never addressed a detailed technical report of a systemic problem
>> in the same. Bad tech, bad customer service, and it looked like this
>> was a more common scenario than among other manufacturers. All of it
>> boiled down to a nasty case of being a bad candidate for spending time
>> and money.
>
> Samsung, uh? Here's my story of today. My fried just bought two external
> USB drives. I wanted to know which brand the HD is, so I checked with
> hdparm -I, and googled for SAMSUNG HD204UI. I found a story about a bug
> which makes the drive sometimes forget to write a block when it is
> attached to a SATA adapter in AHCI mode and when the ATA command
> "IDENTIFY DEVICE" is sent (like in hdparm -I or when using the
> smartmontools). There is a firmware patch for this, this is good. But on
> the annoying side:

That could very likely be the nature of the initial symptom for my
second failed drive. I recall being angry about silent corruption,
with SMART not reporting anything interesting. Drive failed
differently later on, IIRC. I still have it in the same eSATA external
enclosure I was using at the time. I'll have to look.

>
> - You need to make a DOS boot floppy and copy the patch there. I don't
>  know how exactly to do this, and I read about people using Linux who
>  needed over an hour for this or even failed. Can't they just let me
>  download an image I can boot from?

Any idea if it works from FreeDOS?

>
> - It doesn't work over USB, so I would have to install the drive in a PC.

Does the enclosure doens't contain an eSATA port? That's almost
certain to be a direct passthrough.

>
> - The new firmware has exactly the same revision number. How stupid is
>  this?? I cannot even find out whether the drives have the problem or
>  not. Except by trying to reproduce the problem.

Sounds like the only way you can be certain the drives don't have the
problem is by installing the patched firmware.

>
> Here's a link to the but I described, but It's German only.
> http://www.heise.de/ct/meldung/Firmware-Patch-fuer-Samsung-Festplatte-EcoGreen-F4-HD204UI-Update-1150154.html
> I also read some angry comments about Samsung there. Question is, are
> other manufacturers better? And wasn't Samsung Electronics bought by
> Seagate anyway?
>
>
> Any idea whether an external USB drive case might count as a SATA
> controller in AHCI mode? I tried to trigger the bug, but that did not
> happen, so I guess it's fine, at least when being in the USB case.

The drive inside is SATA, so the USB enclosure is translating USB
mass-storage commands to commands the SATA drive can understand.
AFAIK, AHCI vs Legacy mode is a function of the SATA *controller*, not
of the drive itself; legacy mode takes an older protocol, converts it
to SATA commands, and then dispatches those SATA commands. I'll
venture a guess that when going through Legacy mode, whatever commands
trigger the bug aren't used in the Legacy->SATA conversion, whereas
AHCI, with its closer-to-metal nature, exposes those commands for use.
Whether or not the USB enclosure will trigger those bugs would depend
on whether or not the USB Mass Storage<->SATA translation uses those
commands or not.

(In my mind, this is feels like the old AGP->PCI Express transition.
Early PCIe video cards were actually AGP cards with an AGP->PCIe
bridge/adapter chip onboard, because it was faster and cheaper to get
to market by throwing an adapter component in the sequence. However, I
don't know enough about the ASIC market for USB hard drive enclosures
to know whether chips with adapter layers (like that legacy->SATA
command translation, but at a hardware level, making it
USB->legacy->SATA) will be the cheaper part to source than chips which
convert more directly between the USB Mass Storage and SATA
protocols.)


>
> Another problem is that data access frequently stalls on her PC, like when
> transferring data or doing a mke2fs. After a while, this message appears
> in syslog, and the process continues for a while, until it happens again:
>
> usb 1-4: reset high speed USB device using ehci_hcd and address 7

That looks incredibly familiar. I kept getting those, and then
switched to my enclosure's eSATA port. That's when the drive started
giving me different problems. At the time, I'd assumed it was the
enclosure's USB components at fault.

>
> Same problem with a GRML boot cd and on another USB port. Happens with
> both drives. But it is fine on my PC.


It sounds like the drives might be salvageable with a firmware patch,
now. I'd suggest extracting the drives, plugging them into a PC,
updating the firmware, and then putting them back in the enclosure.
That way, you're certain the drives don't still have the known-buggy
firmware, at the expense of some labor.

-- 
:wq

Re: [gentoo-user] Re: Hard drive RPMs and data speed.

[Volker Armin Hemmann]

Am Montag 31 Oktober 2011, 17:21:23 schrieb Alex Schuster:
> Michael Mol writes:
> > My point is that the numbers aren't what mattered here. My point is
> > that SAMSUNG sold me a shoddy product, replaced it with another
> > instance of the the same shoddy product, wouldn't replace it again,
> > and never addressed a detailed technical report of a systemic problem
> > in the same. Bad tech, bad customer service, and it looked like this
> > was a more common scenario than among other manufacturers. All of it
> > boiled down to a nasty case of being a bad candidate for spending time
> > and money.
> 
> Samsung, uh? Here's my story of today. My fried just bought two external
> USB drives. I wanted to know which brand the HD is, so I checked with
> hdparm -I, and googled for SAMSUNG HD204UI. I found a story about a bug
> which makes the drive sometimes forget to write a block when it is
> attached to a SATA adapter in AHCI mode and when the ATA command
> "IDENTIFY DEVICE" is sent (like in hdparm -I or when using the
> smartmontools). There is a firmware patch for this, this is good. But on
> the annoying side:
> 
> - You need to make a DOS boot floppy and copy the patch there.

nope, just use systemrescuecd. It has a freedos boot image (or had.. last time 
I needed it.. was long ago)

> - The new firmware has exactly the same revision number. How stupid is
>   this?? I cannot even find out whether the drives have the problem or
>   not. Except by trying to reproduce the problem.

yes, that is stupid. but you can just run the patch agan.

> 
> Here's a link to the but I described, but It's German only.
> http://www.heise.de/ct/meldung/Firmware-Patch-fuer-Samsung-Festplatte-EcoGre
> en-F4-HD204UI-Update-1150154.html I also read some angry comments about
> Samsung there. Question is, are other manufacturers better? And wasn't
> Samsung Electronics bought by Seagate anyway?
> 

yes

> 
> Any idea whether an external USB drive case might count as a SATA
> controller in AHCI mode? I tried to trigger the bug, but that did not
> happen, so I guess it's fine, at least when being in the USB case.

none. Get esata.

> 
> Another problem is that data access frequently stalls on her PC, like when
> transferring data or doing a mke2fs. After a while, this message appears
> in syslog, and the process continues for a while, until it happens again:
> 
> usb 1-4: reset high speed USB device using ehci_hcd and address 7
> 

your case is crap.

> Same problem with a GRML boot cd and on another USB port. Happens with
> both drives. But it is fine on my PC.

-- 
#163933